Models, products and AI startups, near-duplicates collapsed to the most credible source
McKinsey's Kutcher: AI creates jobs and opportunityEric Kutcher, McKinsey & Co. chair of North America, is very bullish on the US economy, saying there will be more jobs in 2035 than today. He says AI is creating jobs, pointing to high demand for software developers who help AI write code.
TII releases Falcon-ASR Arabic speech recognition modelAbu Dhabi's Technology Innovation Institute released Falcon-ASR, a 1.6B-parameter speech recognition model focused on Arabic and the Emirati dialect, also supporting English, French, Spanish and Portuguese. It achieved an average 20.92% WER across six Arabic test sets, 2.25 points better than the best published result.
Hilt raises $4.2M seed fundingCybersecurity startup Hilt emerged from stealth with $4.2M in seed funding led by Array Ventures, with participation from Verdict Capital, Base10 Partners and others. Including a $500K pre-seed led by Pear VC, total funding reaches $4.7M. Hilt helps companies detect and stop unusual data activity, including sensitive information theft.
Datavault AI Hit with Market-Manipulation Securities SuitShareholders filed a securities class action in the Eastern District of Pennsylvania against Datavault AI and certain executives, alleging promotional announcements around AI, blockchain, quantum computing and data monetization painted an inflated picture of prospects, with claims also touching liquidity, financing and going-concern issues.
Fine-Tuned Nemotron Reaches Gold Level at IOI and IMOStarting from Nemotron 3, teams used supervised fine-tuning, reinforcement learning and an inference loop to build specialists that hit gold-medal level at both IMO 2026 and IOI 2026. The IOI run was an unofficial live benchmark under human-equivalent time and submission limits, while IMO proofs were graded by official IMO graders.
llama.cpp adds local support for GLM 5 Flash MTPA llama.cpp pull request adds support for GLM 5 Flash MTP, allowing users to run the model locally. The update was submitted by contributor pwilkin.
Autodesk goes all in on AI at AU conferenceAt Autodesk University in Las Vegas, some 12,000 AEC professionals gathered as Autodesk unveiled AI features: project intelligence in Forma, Building Layout Explorer in Revit, and the autonomous Autodesk Assistant rolling out in 2027. CEO Andrew Anagnost acknowledged the week was 'very doomy' in the AI world.
Nippan confirms selling books to AnthropicJapanese book wholesaler Nippan Group confirmed in a letter to the Japan Book Publishers Association that subsidiary Nippan IPS struck a deal with Anthropic in November 2024 and sold books from December that year through last December, without disclosing volume or amount. It said it had asked the purpose but did not know the books were for AI learning; the deal is part of Anthropic's Project Panama.
LiquidAI teases new LFM model release todayA Reddit post cites a tweet from a LiquidAI figure hinting that a new LFM model will be released today, asking users what size they want. Details remain scarce; updates are expected on the company's Hugging Face page.
Study: 47.5% of US lawyers report burnout, AI adds pressureThe largest study to date on US lawyer mental health found 47.5% report high or very high burnout, rising to 69.5% for those working 71+ hours a week. Authors cite practice demands, field, and work-life conflict as main factors, and mention AI disruption; lawyers under 30 have over three times the rate of moderate or major depression than those over 61.
OpenAI adds College Planner to ChatGPT for TeensOpenAI is adding College Planner to ChatGPT for Teens to help students manage college applications, alongside new flashcards, quizzes, and a teen AI council.
Google launches Playground experimental gaming platformGoogle launched Playground, an experimental gaming platform that lets users create, play and share custom games via text prompts without coding. It supports multiplayer and leaderboards, and is live at playground.google, with professional-grade tools from Unity Spark coming later. The move aims to lower barriers to game creation.
AI could upend food delivery as DoorDash eyes startup BitesDoorDash processed 970 million orders and $4.5B revenue in Q2, while Bites, a 10-person Bay Area startup with ~300 restaurants, drew its attention this summer. In August DoorDash emailed restaurants warning they may be listed on Bites without consent.
Micron Says NVHBM to Improve ProfitabilityMicron says NVHBM moves the memory controller to the base die, cutting power 15%, boosting memory bandwidth by up to 30%, and reducing I/O PHY package area by up to 67%. The technology is expected to improve profitability even with outsourced base dies.
Kazakhstan religious congress focuses on AI ethics and climateThe 24th Secretariat meeting of the Congress of Leaders of World and Traditional Religions opened in Astana with delegates from about 20 countries, set to review a draft ethical code for AI use in religious contexts and a document on religions' role against climate change.
CoreWeave to Launch First India Data Centers via AdaniConneXAI cloud provider CoreWeave is partnering with Adani Group venture AdaniConneX to build its first data centers in India, continuing its Asia expansion.
Council of Europe and Microsoft Sign AI Cooperation AgreementThe Council of Europe and Microsoft have signed a cooperation agreement on artificial intelligence. Further details of the agreement were not disclosed.
Study: Claude and ChatGPT Give Different Shopping Advice Based on WealthResearch examining how Claude and ChatGPT act as shopping assistants found the chatbots offer different price recommendations based on users' wealth, a finding that could lead to more calls for regulation.
Investors raise the bar: there was never an easy AI eraInvestor Maor Farid argues that in 2021 an AI label was enough to raise money, but many products failed on retention and the market is correcting. Defensible AI businesses must dramatically improve business-critical processes and combine domain expertise with proprietary context, not minor gains on noncritical tasks.
China's Open-Weight AI Models Gain Global Users, but Who Captures Value?China's open-weight AI models are attracting a growing global user base, but it remains unclear who is capturing the economic value from their adoption.
OpenAI's New Agent Aims to Automate Online Tasks, Early Test Shows BugsOpenAI has introduced a new always-on agent designed to automate online tasks like buying furniture. In an initial test, the agent was buggy, failed a captcha, and responded in an overly familiar way.
Microsoft, Nvidia CEOs to unveil new AI laptop in San FranciscoMicrosoft and Nvidia CEOs will jointly unveil a new AI laptop at a San Francisco event, signaling continued deepening of the two firms' cooperation in AI PCs.
Nvidia bets these startups will lift its stockBarron's reports that Nvidia is betting on a set of startups in the hope of lifting its own stock, without further specifics.
Ex-researcher claims China has spies at Anthropic and OpenAINDTV reports a former researcher's claim that China already has spies at Anthropic and OpenAI, without further verifiable details.
AI Bots Swarm Fed's FRED Data Site as Dimon Warns of Security RisksAI agents have been swarming the Federal Reserve's FRED data site, raising concerns about potential errors in how economic information is described and cited. JPMorgan CEO Jamie Dimon also warned about AI security risks.
Poll: Most US Voters Say Trump, Congress Don't Take AI Risks SeriouslyA Reuters/Ipsos poll found that most US voters believe President Trump and Congress are not taking the risks posed by artificial intelligence seriously.
Information Workers Launches Enterprise AI Governance Practice in USConsultancy Information Workers has launched a US-wide enterprise AI governance practice. It combines auditable data architectures, compliance sandboxes, and a step-by-step AI methodology to help companies adopt AI securely and eliminate shadow AI.
Discussion: why people dislike AI codingAn article and Hacker News discussion rebut common objections to AI coding: software is valued for outcomes rather than the act of creation, developers are paid to deliver valuable software rather than type code, and greater efficiency tends to increase demand.
University of Tennessee Launches Institute for the Future of BusinessThe University of Tennessee at Knoxville and its Haslam College of Business have launched the Institute for the Future of Business to help students and employers adapt to workplace changes driven by AI and other technologies. It will partner with Oak Ridge Associated Universities to study AI's impact on work in Tennessee.
ChatGPT for Teens Is Not Safe for Kids, Report SaysA report from Common Sense Media's Youth AI Safety Institute found that ChatGPT for Teens often failed to promptly alert parents, or did not alert them at all, when teen accounts discussed suicide, self-harm or eating disorders. It also did not reliably suggest crisis hotlines or professional help. The mode has been the default for users aged 13 to 17 since August.
The A.I. Daily